Original paper published by Satoshi Nakamoto, compares between one-cpu-one-vote and one-ip-address-one-vote, then chooses former because of the reason that someone would be able to allocate more than one IP.

This statement also sounds against anyone with huge hash power.

With this, it seems that, GPU / FPGA / ASIC hashing was not predicted at the time of publishing this paper.

Is this an important change, or just a design decision for the growth of Bitcoin ecosystem? What's the motivation behind one-cpu-one-vote rule? And, will it still applies to Bitcoin ecosystem?

  • We're already basically out of IP addresses. Some parties got tons of addresses, others got few. And give monetary value to them and that is something sure to make them positively scarce. Dec 7, 2012 at 23:02

3 Answers 3


The motivation for this rule is simple - there is a requirement to have some voting mechanism , and this 1-CPU-1-vote mechanism is fair, in the sense that everyone can buy more CPU (or GPUs or ASICs) and mine. The rewards are proportional to the invested effort - better harder costs more, it doesn't matter if it's CPUs, GPUs or ASICs.

So, the protocol rewards miners in proportion to the amount of money they invest in buying mining gear. It's certainly better than rewarding per IP address, which is controlled by various organizations and ultimately centrally managed.

Also, I believe satoshi did foresee the evolution in mining gear, although it seems that it's not mentioned in the original whitepaper.


First, I'd like to point out that hash power in Bitcoin can indeed be regarded as voting power, but only about the part that is absolutely necessary. Bitcoin is generally a zero-trust system, which means all data received from the network is considered untrusted and verified. Rules like rate of currency introduction, not allowing a coin to be spent twice, requirement of valid signatures, ... are never voted about. Even a 51% of mining power cannot change these rules, as they would only get themselves ignored by the rest of the network.

The one thing they vote about is the ordering of otherwise valid transactions. This is necessary as physics imposes minimum times for communication across the globe, so in case there are two attempts to spend the same coin, the whole world cannot be expected to agree about which was first.

That said, the choice for using "one CPU per vote" was probably indeed made without considering the evolution to GPUs, FPGAs and ASICs. Originally, Satoshi envisioned every full node on the network to be a miner, and even asked to avoid the arms race to more efficient mining hardware as long as possible.

Still, I believe assigning votes to computational power (proof of work) makes sense. It may not favor equality (those with better hardware get more), but it is fair (everyone can invest in hardware and reap the rewards). Except for systems which depend on data in the system itself (proof of stake, proof of activity), I don't know anything that is as impossible to fake or circumvent as proof-of-work.

  • 1
    AFAIK satoshi did anticipate mining using GPUs and custom hardware.
    – ripper234
    Dec 7, 2012 at 12:34
  • 1
    I don't find it reasonable that Satoshi did not consider GPUs and ASICs. He did not acknowledge them in his paper, but I guess he just made a first-order approximation for simplicity. His request for a gentleman's agreement doesn't seem to indicate concern about the eventual advent of these, but rather tactical considerations with respect to its timing. On the other hand, that request only came after GPU miners were being developed, so I don't know of any definite proof (other than knowledge of the person) that he considered them in advance. Dec 7, 2012 at 12:35

The motivation is economic. Bitcoin is an instrument for free market global trade. As such, it should not be able to be censored.

From the whitepaper abstract:

The longest chain not only serves as proof of the sequence of events witnessed, but proof that it came from the largest pool of CPU power.

Satoshi acknowledges competing large pools of CPU power. I don't think the form of the computers matters. The largest pool simply has the most votes. The reason for that is the same reasons that motivate global free trade policies in general, namely: Honest competition produces better products, in this case a stronger network capable of serving the purpose of censorship-resistant digital cash.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.